Conventionally, giving ruby characters alongside a character string included in a document is generally performed. The ruby characters are annotation showing reading and meaning of the character string, and to be arranged near the character string so that a correspondence relation with the character string is understood (for example, see Japanese Laid-Open Patent Publication No. 2007-72594 and Japanese Laid-Open Patent Publication No. 2009-53838).
Generally, the size of a ruby character is decided depending on the size of a character and the size of a line space of an original document. In this case, even though the sizes of characters and the sizes of line spaces are similar to each other in the original document, the size of the ruby character is changed unless they are the same so that the document comes to lack unity.
Therefore, the technique has been proposed that a document is divided into a plurality of areas to classify areas in which the sizes of characters and the sizes of line spaces are similar to each other as a same group, and character sizes of a translated sentence that is given alongside a character string are unified among areas included in the same group (for example, see Japanese Laid-Open Patent Publication No. 2009-193283 and Japanese Laid-Open Patent Publication No. 2010-128599).
However, in the case of a document having a less difference among character sizes and line spaces in an original document in which it is better to unify character sizes of a translated sentence in the whole document, and a document in which various sizes of characters are used on purpose so as to appear to give a translated sentence in each size corresponding to each character size, the related art in Japanese Laid-Open Patent Publication No. 2009-193283 and Japanese Laid-Open Patent Publication No. 2010-128599 has a possibility that the character sizes of the translated sentence are not only unified in the former, but also in the latter, the character sizes of the translated sentence are unified among areas where the character sizes are similar to each other, so that it is impossible to respond to a request such that the character sizes of the translated sentence are not unified in the latter.